The sunk cost fallacy: when is it time to step away from a project?


The sunk cost fallacy compels us to pursue projects that we’re better off abandoning. Take these steps to prevent it surfacing in your workplace.

Kathy Hannun was working at X, a secretive research and development company owned by Alphabet, when a revolutionary idea was brought to her attention.

“The pitch was: let’s create replacements for fossil fuels using carbon and hydrogen atoms sourced from nature, so that when the fuels are combusted, there are no net carbon emissions,” says Hannun.

These green fuels would be derived from an abundant resource: seawater. And they would be so similar to fossil fuels that no new vehicle technology would be required. 

As a ‘rapid evaluator’ at X, it was Hannun’s job to screen incoming development proposals like this and decide which ones were worthy of funding. 

She quickly green-lit the fuel proposal and christened it Project Foghorn.

Initially, things looked promising. The Project Foghorn team successfully broke down seawater into its constituent parts and managed to produce viable fuel.

When she started as rapid evaluator at X, she had consulted with stakeholders in an attempt to become a more astute decision-maker. In the process, she learned about the sunk cost fallacy: in simple terms, realising you’ve made a poor decision, but continuing with the same course of action regardless. 

Hannun knew the nature of the work conducted at X would make the company susceptible to the sunk cost fallacy. So she decided to put a ‘kill signal’ in place at the start of each project she approved, so she’d know when to pull the plug.

The kill signal for Project Foghorn was reaching the conclusion that fuel could not be produced for less than $5 a gallon. 

Hannun and her team kept getting close to the $5 mark, but eventually the data showed them that the technology would not enable them to achieve that price.  

Hannun, who went on to co-found the geothermal heating and cooling startup Dandelion energy, was tempted to persist, but she knew the data wasn’t lying.

She avoided the sunk cost fallacy, sometimes known as escalation bias, by abandoning the project, and says being aware of the phenomenon has made her a better leader.

But awareness is not always enough. Researchers say the sunk cost fallacy can be fiendishly difficult to resist, even if you recognise it.

Irrational drivers contribute to the sunk cost fallacy

Once you start looking, you find examples of the sunk cost fallacy everywhere. 

At work, for example, you might continue to pay for a failing performance management system simply because it was your idea to implement it, or continue running an ineffective employee engagement program because setting it up took time and money. 

In leadership situations, the sunk cost fallacy can encourage us to double down on bad policies we’ve developed or remain engaged in losing battles – figurative or literal. America’s protracted engagement in Vietnam in the 1970s is a prime example.

The phenomenon manifests in our personal lives, too – for example, convincing ourselves to remain in unhealthy relationships because we’ve already devoted years to them.

If we thought about these situations rationally, we would conclude that continuing in the same manner was a bad idea.

But human behaviour is not always rational, as University of Queensland psychology lecturer Ryan Metcalfe explains.

“If we were to spend our lives trying to systematically analyse every little decision, we would never get anywhere,” he says. “So we tend to take shortcuts, which are known as heuristics,” says Metcalfe.

Photo: George Milton, Pexels

The sunk cost fallacy is similar to the heuristic known as commitment bias, which makes us prone to continue supporting past decisions despite new evidence that casts doubt on them.

The main difference is that the sunk cost fallacy involves past decisions that have cost us money, time, energy or resources that can’t be recovered. 

Why do we behave this way? Professor Barry M. Staw, an organisational behaviour specialist at the University of California Berkeley, says it’s because we have a fundamental aversion to losing – even if that loss appears to be inevitable.

He provides a poignant example.

“I remember my mother declining with cancer and having one treatment after another: incredibly miserable and painful treatments.” 

The situation was grim, but Staw’s father pushed the doctors to keep going. 

“He would say, ‘Let’s just wait for one more procedure.’ And when that procedure failed he would insist that the next procedure would turn things around. This went on for months. My mother was pleading for him to finally let her die.”

In day-to-day life, heuristics often serve us well. But our brains have not yet figured out how to override them when we are faced with complex, high-stakes decisions.

“If you go to law school and absolutely hate it, it’s relatively easy to withdraw at the beginning before too much is invested,” says Staw. “The problem is that most of the big decisions in life are not so obvious.” 

And if you’re three years into a five-year degree, it becomes much harder to detach yourself from the decision. So often, we plough on.

Hannun says HR itself must be careful not to fall prey to the sunk cost fallacy – especially in the current climate. As businesses get back on their feet after the pandemic, many employers are looking to HR for ideas and are increasing HR budgets so new initiatives can be implemented. 

That creates pressure to deliver results, while also raising the stakes.

“The stories of failed startups often feature a founder or team that is given more money than is strictly necessary at the outset,” says Hannun. “A larger budget means you can pursue ideas for longer, perpetuating the sunk cost fallacy.”

The takeaway for HR: as you help your businesses to bounce back from the losses and shocks of the pandemic, don’t be afraid to try new things, but it’s equally important to not be afraid to pull the plug.


Knowing when to call it quits on a project is just one of the critical skills of an effective team. Learn more with AHRI’s short course on Creating High Performance Teams.
Book in for the next course on 17 March.


System override

You might assume that individuals with superior intrinsic intelligence (that is, high IQ and EQ) would find it easier to avoid falling prey to the sunk cost fallacy. But researchers have discovered otherwise.

A recent study by a trio of US academics found that learned knowledge (known as ‘crystallised intelligence’) rather than raw computational power (‘fluid intelligence’) is the key to overcoming the phenomenon.

Put simply, understanding your relevant subject matter counts for more than being naturally clever.

With this in mind, providing specific training at work about behavioural psychology can be useful, says Metcalfe.

“Workers should be trained to recognise situations where they might be susceptible to this, so they can choose to be more systematic from the outset.”

He also supports Hannun’s tactic of assigning a kill signal to projects of significance and says it should be coupled with regular assessment.

“Set up a monthly ‘check in’ with yourself to assess the venture rationally and check for the signal,” he says. “Do this regardless of the investment you’ve already made.”

“People will call it ‘Jim’s baby’ or ‘Mary’s project’… And then that employee has to make sure it’s a success. If it isn’t, they may be out of a job because they’ve become so identified with it.”Professor Barry M. Staw, organisational behaviour specialist, University of California Berkeley

Staw says employers should encourage their employees to consult widely about decisions.

“Ask third parties,” he says. “Give them the data and descriptions of the work situation and ask them what they would do. When you take it away from the individual who has already been sucked in, you get a much more rational decision.”

Many banks already do this when assessing problematic loans, he says.

“When a bad loan exceeds a certain amount of losses, it’s moved to what’s called a work-out group.

“The people in that group don’t care how much money was lost previously. They look at it in an objective way. They concentrate on future outcomes, not the past.”

Something to prove

Company culture can also make the sunk cost fallacy harder to avoid. 

“A big problem at work is the inclination to try to prove to others that you’re a good decision-maker,” says Staw.

This often happens when employees become associated with a particular initiative.

“People will call it ‘Jim’s baby’ or ‘Mary’s project’ or similar,” says Staw. “And then that employee has to make sure it’s a success. If it isn’t, they may be out of a job because they’ve become so identified with it.”

String connecting a balloon and weight is cut with scissors to represent the sunk cost fallacycissors

In these situations, it’s almost rational for employees to persist beyond where they should. The solution is in enhancing psychological safety so people aren’t so fearful of failure, says Staw.

“There are two kinds of accountability,” he explains. “Usually, we think of accountability in terms of outcomes: people are rewarded on the basis of whether their project or performance or department succeeds.”

That approach can be useful, but in isolation it can be counter-productive. 

“In addition to outcome accountability, it’s important to have process accountability,” says Staw. “That means rewarding people if they’ve gone through a conscientious decision-making process. They need to know that they can still be considered a good employee if things just don’t work out.”

guest
0 Comments
Inline Feedbacks
View all comments
More on HRM
Sorry, no posts matched your criteria.

The sunk cost fallacy: when is it time to step away from a project?


The sunk cost fallacy compels us to pursue projects that we’re better off abandoning. Take these steps to prevent it surfacing in your workplace.

Kathy Hannun was working at X, a secretive research and development company owned by Alphabet, when a revolutionary idea was brought to her attention.

“The pitch was: let’s create replacements for fossil fuels using carbon and hydrogen atoms sourced from nature, so that when the fuels are combusted, there are no net carbon emissions,” says Hannun.

These green fuels would be derived from an abundant resource: seawater. And they would be so similar to fossil fuels that no new vehicle technology would be required. 

As a ‘rapid evaluator’ at X, it was Hannun’s job to screen incoming development proposals like this and decide which ones were worthy of funding. 

She quickly green-lit the fuel proposal and christened it Project Foghorn.

Initially, things looked promising. The Project Foghorn team successfully broke down seawater into its constituent parts and managed to produce viable fuel.

When she started as rapid evaluator at X, she had consulted with stakeholders in an attempt to become a more astute decision-maker. In the process, she learned about the sunk cost fallacy: in simple terms, realising you’ve made a poor decision, but continuing with the same course of action regardless. 

Hannun knew the nature of the work conducted at X would make the company susceptible to the sunk cost fallacy. So she decided to put a ‘kill signal’ in place at the start of each project she approved, so she’d know when to pull the plug.

The kill signal for Project Foghorn was reaching the conclusion that fuel could not be produced for less than $5 a gallon. 

Hannun and her team kept getting close to the $5 mark, but eventually the data showed them that the technology would not enable them to achieve that price.  

Hannun, who went on to co-found the geothermal heating and cooling startup Dandelion energy, was tempted to persist, but she knew the data wasn’t lying.

She avoided the sunk cost fallacy, sometimes known as escalation bias, by abandoning the project, and says being aware of the phenomenon has made her a better leader.

But awareness is not always enough. Researchers say the sunk cost fallacy can be fiendishly difficult to resist, even if you recognise it.

Irrational drivers contribute to the sunk cost fallacy

Once you start looking, you find examples of the sunk cost fallacy everywhere. 

At work, for example, you might continue to pay for a failing performance management system simply because it was your idea to implement it, or continue running an ineffective employee engagement program because setting it up took time and money. 

In leadership situations, the sunk cost fallacy can encourage us to double down on bad policies we’ve developed or remain engaged in losing battles – figurative or literal. America’s protracted engagement in Vietnam in the 1970s is a prime example.

The phenomenon manifests in our personal lives, too – for example, convincing ourselves to remain in unhealthy relationships because we’ve already devoted years to them.

If we thought about these situations rationally, we would conclude that continuing in the same manner was a bad idea.

But human behaviour is not always rational, as University of Queensland psychology lecturer Ryan Metcalfe explains.

“If we were to spend our lives trying to systematically analyse every little decision, we would never get anywhere,” he says. “So we tend to take shortcuts, which are known as heuristics,” says Metcalfe.

Photo: George Milton, Pexels

The sunk cost fallacy is similar to the heuristic known as commitment bias, which makes us prone to continue supporting past decisions despite new evidence that casts doubt on them.

The main difference is that the sunk cost fallacy involves past decisions that have cost us money, time, energy or resources that can’t be recovered. 

Why do we behave this way? Professor Barry M. Staw, an organisational behaviour specialist at the University of California Berkeley, says it’s because we have a fundamental aversion to losing – even if that loss appears to be inevitable.

He provides a poignant example.

“I remember my mother declining with cancer and having one treatment after another: incredibly miserable and painful treatments.” 

The situation was grim, but Staw’s father pushed the doctors to keep going. 

“He would say, ‘Let’s just wait for one more procedure.’ And when that procedure failed he would insist that the next procedure would turn things around. This went on for months. My mother was pleading for him to finally let her die.”

In day-to-day life, heuristics often serve us well. But our brains have not yet figured out how to override them when we are faced with complex, high-stakes decisions.

“If you go to law school and absolutely hate it, it’s relatively easy to withdraw at the beginning before too much is invested,” says Staw. “The problem is that most of the big decisions in life are not so obvious.” 

And if you’re three years into a five-year degree, it becomes much harder to detach yourself from the decision. So often, we plough on.

Hannun says HR itself must be careful not to fall prey to the sunk cost fallacy – especially in the current climate. As businesses get back on their feet after the pandemic, many employers are looking to HR for ideas and are increasing HR budgets so new initiatives can be implemented. 

That creates pressure to deliver results, while also raising the stakes.

“The stories of failed startups often feature a founder or team that is given more money than is strictly necessary at the outset,” says Hannun. “A larger budget means you can pursue ideas for longer, perpetuating the sunk cost fallacy.”

The takeaway for HR: as you help your businesses to bounce back from the losses and shocks of the pandemic, don’t be afraid to try new things, but it’s equally important to not be afraid to pull the plug.


Knowing when to call it quits on a project is just one of the critical skills of an effective team. Learn more with AHRI’s short course on Creating High Performance Teams.
Book in for the next course on 17 March.


System override

You might assume that individuals with superior intrinsic intelligence (that is, high IQ and EQ) would find it easier to avoid falling prey to the sunk cost fallacy. But researchers have discovered otherwise.

A recent study by a trio of US academics found that learned knowledge (known as ‘crystallised intelligence’) rather than raw computational power (‘fluid intelligence’) is the key to overcoming the phenomenon.

Put simply, understanding your relevant subject matter counts for more than being naturally clever.

With this in mind, providing specific training at work about behavioural psychology can be useful, says Metcalfe.

“Workers should be trained to recognise situations where they might be susceptible to this, so they can choose to be more systematic from the outset.”

He also supports Hannun’s tactic of assigning a kill signal to projects of significance and says it should be coupled with regular assessment.

“Set up a monthly ‘check in’ with yourself to assess the venture rationally and check for the signal,” he says. “Do this regardless of the investment you’ve already made.”

“People will call it ‘Jim’s baby’ or ‘Mary’s project’… And then that employee has to make sure it’s a success. If it isn’t, they may be out of a job because they’ve become so identified with it.”Professor Barry M. Staw, organisational behaviour specialist, University of California Berkeley

Staw says employers should encourage their employees to consult widely about decisions.

“Ask third parties,” he says. “Give them the data and descriptions of the work situation and ask them what they would do. When you take it away from the individual who has already been sucked in, you get a much more rational decision.”

Many banks already do this when assessing problematic loans, he says.

“When a bad loan exceeds a certain amount of losses, it’s moved to what’s called a work-out group.

“The people in that group don’t care how much money was lost previously. They look at it in an objective way. They concentrate on future outcomes, not the past.”

Something to prove

Company culture can also make the sunk cost fallacy harder to avoid. 

“A big problem at work is the inclination to try to prove to others that you’re a good decision-maker,” says Staw.

This often happens when employees become associated with a particular initiative.

“People will call it ‘Jim’s baby’ or ‘Mary’s project’ or similar,” says Staw. “And then that employee has to make sure it’s a success. If it isn’t, they may be out of a job because they’ve become so identified with it.”

String connecting a balloon and weight is cut with scissors to represent the sunk cost fallacycissors

In these situations, it’s almost rational for employees to persist beyond where they should. The solution is in enhancing psychological safety so people aren’t so fearful of failure, says Staw.

“There are two kinds of accountability,” he explains. “Usually, we think of accountability in terms of outcomes: people are rewarded on the basis of whether their project or performance or department succeeds.”

That approach can be useful, but in isolation it can be counter-productive. 

“In addition to outcome accountability, it’s important to have process accountability,” says Staw. “That means rewarding people if they’ve gone through a conscientious decision-making process. They need to know that they can still be considered a good employee if things just don’t work out.”

guest
0 Comments
Inline Feedbacks
View all comments
Sorry, no posts matched your criteria.
More on HRM